| Name | Version | Summary | date |
| llm-instrumentation |
1.2.11 |
A high-performance instrumentation framework for capturing, streaming, and analyzing internal activations of large language models |
2025-11-03 19:14:19 |
| nnsight |
0.5.10 |
Package for interpreting and manipulating the internals of deep learning models. |
2025-10-30 18:34:08 |
| signxai2 |
0.15.0 |
Extended variant of ZeNNit with SIGNed explanations - Unveiling relevant features by reducing bias |
2025-10-30 11:30:55 |
| fractal-attention-analysis |
1.0.0 |
Fractal-Attention Analysis (FAA) Framework for LLM Interpretability using Golden Ratio Transformations |
2025-10-26 00:09:48 |
| lbmd-sota |
1.0.1 |
Latent Boundary Manifold Decomposition for State-of-the-Art Mechanistic Interpretability |
2025-10-23 21:05:17 |
| nnez |
0.1.3 |
Neural Network Easy Extraction - Simple LLM activation extraction in one line |
2025-10-23 19:54:45 |
| interp-embed |
0.1.0 |
Toolkit for analyzing unstructured datasets with sparse autoencoders |
2025-10-23 05:38:15 |
| shap-enhanced |
0.0.1 |
SHAP-style explainability with enhanced methods for sequential and tabular models |
2025-10-06 09:35:17 |
| verskyt |
0.2.4 |
Python library for Tversky Neural Networks with research tools for model introspection and prototype analysis |
2025-09-02 03:43:03 |
| coret |
0.1.1 |
Surrogate based Concept Retrieval for Large Datasets |
2025-08-06 11:52:02 |
| corelay |
1.0.0 |
CoRelAy is a tool to compose small-scale (single-machine) analysis pipelines to generate analysis data which can then be visualized using ViRelAy. |
2025-07-21 09:48:39 |
| effector |
0.1.12 |
An eXplainable AI package for tabular data. |
2025-07-12 21:11:14 |
| minicons |
0.3.14 |
A package of useful functions to analyze transformer based language models. |
2025-02-20 20:00:59 |
| graphpatch |
0.2.3 |
graphpatch is a library for activation patching on PyTorch neural network models. |
2024-10-19 01:30:59 |
| scanvi-explainer |
0.3.1 |
Exapliner of scANVI using SHAP |
2024-09-27 22:20:01 |
| jax-relax |
0.2.8 |
JAX-based Recourse Explanation Library |
2024-08-29 02:26:43 |
| inseq |
0.6.0 |
Interpretability for Sequence Generation Models 🔍 |
2024-04-13 13:37:37 |